Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Study On Steel Fibre Used in Tunnel Segmental Lining

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Study on Steel Fibre Used in Tunnel Segmental Lining

R. Gong, Maunsell Australia Pty Ltd, N. Gupta, Maunsell Australia Pty Ltd, J. Sekula, Maunsell Australia
Pty Ltd, G. Charlesworth, Maunsell Australia Pty Ltd.

ABSTRACT: Steel fibre has generally been used for shotcrete in temporary works. This paper dis-
cusses the use of SFRC in tunnel segmental lining. Steel fibre reinforced concrete has advantages over
conventional steel bar reinforced concrete in terms of improved structural strength, and its superior
resistance to cracking and crack propagation. Steel fibre can be used alone, or in conjunction with
conventional steel bar, to reinforce plain concrete. The loading conditions and durability require-
ments for tunnels make this composite material particularly attractive for the lining design. The struc-
tural capacity of SFRC lining segment was evaluated with the effect of steel fibre allowed for, based
on which comparisons were made between plain concrete, SFRC, conventional reinforced concrete
and steel fibre + steel bar reinforced concrete.

1 INTRODUCTION

Steel fibre reinforced concrete (SFRC) is a composite material of hydraulic cements, fine and/or
coarse aggregates with discrete steel fibres randomly dispersed throughout the matrix. The fibres act
as “miniature reinforcing bars” to reinforce and restrain micro-cracking essentially, thus improve duc-
tility and post-crack performance of concrete. Unlike wire mesh or rebar, steel fibres provide three-
dimension reinforcement throughout the entire concrete matrix.

SFRC is not a new material in the tunneling world. For years this material has been used for shotcrete
in the mining as well as civil industries, mainly for temporary works. In recent years SFRC has also
been used for permanent segmental lining in a few tunnelling projects such as Channel Tunnel in UK,
the Second Heinenoord Tunnel in Netherlands (Kooiman et al. 1998) and Gold Coast Desalination
Plant in Australia. Steel fibre can also be used in conjunction with conventional steel bar as rein-
forcement for lining segments. One such example is Barcelona Metro Line 9 in Spain, in which the
quantity of conventional steel bar was reduced by the addition of steel fibre. However, use of SFRC in
the reinforcement of tunnel segments is still in the relative early stages, with the design approaches
varying with designers and projects. Lack of generally accepted design code has in some extent hin-
dered the application of this material.

In this paper the advantages and constrains of SFRC used in tunnel segmental lining are discussed.
Structural capacity of lining segment is evaluated with the effect of steel fibre taken into account.
Comparisons are made between plain concrete, steel fibre and steel bar reinforced concrete, and pa-
rametric studies carried out.

2 ADVANTAGES AND CONSTRAINTS

2.1 Advantages

The advantages of SFRC over conventional steel bar reinforced concrete (SBRC) for tunnel segmental
lining are as follows.

1
• Durability

Both SFRC and SBRC structural member may corrode. However, being typically much thinner than
steel bar, steel fibres are not continuous nor connected within concrete matrix, so they provide no
mechanism for propagation of corrosion. Corrosion is limited, thus improving the durability of the
structure.

• Cracking

Micro cracking of concrete matrix occurs in various directions at very small loads or due to natural
shrinkage. Steel fibres distributes throughout the matrix and serve as small reinforcing bars bridging
such cracks, thus improves the material resistance to cracking and crack propagation. The improved
ductility and post-crack performance of SFRC has been proven by numerous tests (Moyson 1994).

• Spalling

Spalling or chipping at edges and corners is a problem for conventional RC segments due to cover re-
quirement and lack of reinforcement in these locations. Steel fibres distribute evenly throughout the
concrete matrix, effectively minimize such spalling.

• Cost

Compared with conventional RC, the use of SFRC expedites fabrication of tunnel lining segments, re-
ducing execution time and saving labour costs. In addition, construction costs may be reduced through
reduced damage during transportation and installation.

2.2 Constraints

Major constraints for use of SFRC in tunnel structures are listed as below.

• Design approach

Conventional RC structure design has a relatively long history, and mature design approach with in-
ternationally accepted design codes available in most countries. Comparatively, SFRC design is still
in its infancy. Due to lack of recognized codes or standards, engineers have usually carried out SFRC
design by adopting the same rules that are used in conventional RC design. However, the ductility and
post-cracking behaviour of concrete is dramatically improved with the addition of steel fibres, and this
can be exploited in the design.

• Ultimate capacity

It is recognized that steel fibre does not significantly improve the ultimate bending or tensile capacity
of concrete, therefore it cannot be a replacement for structural steel. In the loading conditions with
high bending or tension, sufficient continuous steel reinforcement is required to take the full tensile
stress.

• Experience and quality control

The available greater experience in the design, manufacture and use of conventional steel bar rein-
forced segments provides confidence of product quality. In contrast, the lack of experience in SFRC
segment requires strict quality control on manufacturing as well as comprehensive testing regime to
establish that the designer’s structural criteria have been achieved.

3 DESIGN OF STEEL FIBRE REINFORCED CONCRETE

3.1 Material property

2 International Conference on Deep Excavations (ICDE 2008)


In this paper the material property for SFRC is defined according to the German recommendations
(German Concrete Association, 1992), where the stress-strain relation is established by means of four-
point bending test. A typical experimental load-deflection curve for SFRC beam is shown in Figure 1.

Figure 1. Typical load-deflection curve of four point bending test on a SFRC beam

As shown in Figure 1, three characteristic deflections, δ1, δ2 and δ3, on the curve are defined by δ1 =
the deflection corresponding to the maximum value of the load Fu within the 0.1mm interval, δ2 = δ1 +
0.65 mm and δ3 = δ1 + 3.15 mm.

First crack flexural strength, fct1, and post-cracking equivalent flexural strengths, fct2 and fct3, corre-
sponding to deflections δ1, δ2 and δ3 respectively, are defined by

F L
f = u (1)
ct1
bh 2

D L
f ,2
f = (2)
ct 2
0.5bh 2

D L
f ,3
f = (3)
ct 3
3.0bh 2

where Fu = first cracking force; L = beam span; b = beam width; h = beam height; Df,2 = crack tough-
ness corresponding to δ2, i.e. the area below the load-deflection graph until δ2; and Df,3 = crack tough-
ness corresponding to δ3.

Characteristic flexural strengths are then obtained from

f ct1,c = (0.8 − α ) ⋅ k ⋅ f (4)


ct1

f ct 2,c = 0.37 ⋅ υ ⋅ k ⋅ f (5)


ct 2

f ct 3,c = 0.37 ⋅ υ ⋅ k ⋅ f (6)


ct 3

where α represents the existence of normal force in the actual situation (= 0 in the case of compres-
sive normal force or 0.25 in the case of tensile or missing normal force); k allows for the discrepancy
between the preliminary test and the quality test (= 0.9 for simplication); ν allows for the long term

International Conference on Deep Excavations (ICDE 2008) 3


behaviour and size effect, and varies from 0.85 to 0.6 with structure thickness increases from 150mm
to 300mm; and the factor 0.37 represents the difference of section modulus between cracked and un-
cracked cross-section based on the assumption of a compression zone equal to 0.1 times the structure
height after cracking.

The characteristic stress-strain relation is shown in Figure 2, where f’c = characteristic compressive
cylinder strength of concrete, εt1 = first crack tensile strain, εtu = ultimate tensile strain, εc1 = end of the
elastic compressive strain, and εcu = ultimate compressive strain.

Figure 2. Characteristic stress-strain relation of SFRC

3.2 Safety factors

Based on the ultimate limit state design and partial safety factor theory, the characteristic material
strength is divided by a material factor to give the design strength. Material factors vary with codes
and materials. In this paper material factors of 2.0 are suggested for concrete in compression, 1.0 for
concrete in tension, and 1.25 for steel reinforcement.

3.3 M-N Interaction diagram

Bending moment (M) and axial force (N) interaction diagram can be constructed with the same proce-
dure as that for conventional RC. It is assumed that plane section of a structural member remains
plane after straining, so distribution of strains across the section is linear. Stress across the section is
derived from the stress-strain relation shown in Figure 2, in which rectangular parabolic stress block is

Figure 3. Typical M-N interaction diagram for SFRC

4 International Conference on Deep Excavations (ICDE 2008)


assumed in compression zone and two-stage linear stress distributions in tension zone. For an assumed
neutral axis depth M and N can be determined with integration of stresses over the cross-section, and
M-N interaction diagram constructed by varying neutral axis depth. A spreadsheet was developed for
this purpose, and a typical M-N interaction diagram for SFRC is shown in Figure 3.

For steel fibre used in conjunction with conventional steel bar as reinforcement for concrete
(SFSBRC), M-N interaction diagram can be constructed by taking into account the contribution of
steel bar in the integration of stress. Stress in the steel bar is derived from normal stress-strain relation
of steel.

4 COMPARISON BETWEEN PLAIN CONCRETE, SFRC, SBRC AND SFSBRC

Effect of steel fibre on the capacity of structure member was studied by comparing the M-N interac-
tion diagrams of plain concrete, SFRC, SBRC and SFSBRC tunnel segmental lining. A cross-section
of 1000mm x 400mm was used in the analysis. Typical material parameters of SFRC for Dramix® RC-
80/60-BN with dosage of 30 kg/m3 are as follows: f’c = 50MPa, fct1 = 5.8 MPa, fct2 = 4.1 MPa, and fct3
= 3.6 MPa, εt1 = 0.0001, εtu = 0.01, εc1 = 0.002, εcu = 0.005. For SBRC and SFSBRC, steel bars were
assumed to arrange symmetrically on two faces, and the total area (As) was 0.4% of the cross-section
area (Ac). The M-N interaction diagrams are shown in Figure 4.

Figure 4. Comparison of structural capacity for plain concrete, SFRC, SBRC and SFSBRC

From the figure it can be seen plain concrete has no bending capacity in the case of pure bending, i.e.
bending without axial compressive force. With the reinforcement of steel fibre, concrete can resist
pure bending. Steel fibre generally enhances resistance of a concrete member to bending instead of
compression. That is, in the lower portion of the interaction diagram where failure is dominated by
bending, structural capacity is increased by the addition of steel fibre. For the upper portion where
compression failure dominates, structural capacity is the same for SFRC and plain concrete. There-
fore, the effectiveness of steel fibre to increase structural capacity depends on the loading condition of
the member. The larger the eccentricity of the loading, the larger the enhanced capacity.

Steel bar increases concrete’s capacity in both bending- and compression-dominated failures, as can
be seen in the figure. With the addition of steel fibre, resistance to bending will be further increased,
but there is no change on the compressive capacity. The effect of steel fibre on SBRC is basically
similar to that on plain concrete. In the situation of steel fibre used together with steel bar, quantity of
steel bar may be reasonably reduced for the load case of large eccentricity considering the contribu-
tion of steel fibre. In Figure 5 a conventional SBRC segment with 0.6% steel bar is compared with a
SFSBRC segment with 0.4% steel bar. Material parameters for concrete are the same as the above
case. The two segments exhibit similar structural capacity in the bending-dominated failure zone.

Once in place tunnel lining segments in service are generally subject to high compressive loads. In
this case the contribution of steel fibre may not be significant. However, during handling, transporta-

International Conference on Deep Excavations (ICDE 2008) 5


tion and installation, tunnel lining segments must resist bending without compression. The enhance-
ment of steel fibre on the bending resistance can be employed for those situations. Use of SFRC, or
SFSBRC with reduced steel bar quantity, may be justified for this case.

Figure 5. Comparison of structural capacity for SBRC and SFSBRC with reduced steel bar quantity

5 PARAMETRIC STUDIES

5.1 Influence of flexural strength

As indicated the major difference between plain concrete and SFRC is the post-cracking behaviour,
based on which flexural strengths of fct1, fct2 and fct3 are defined for SFRC. Flexural strengths are re-
lated to the dosage of steel fibre in concrete, and they can be obtained from technical manuals for re-
spective products. Theoretical analyses were carried out to study the effect of the flexural strength pa-
rameters on the structural capacity of SFRC, and the results are shown in Figures 6, 7 and 8. A cross-
section of 1000mm x 400mm was assumed for these analyses, with f’c = 50 MPa and other major ma-
terial parameters indicated in the figures.

Figure 6. Influence of fct1

From Figure 6 it can be seen that fct1 has little effect on the M-N interaction diagram. fct1 may be asso-
ciated with the loading under which the first crack appears, but it does not contribute to the structural
capacity at ultimate limit state where post-cracking behaviour dominates.

Figures 7 and 8 show the effect of post-cracking flexural strengths, fct2 and fct3, on the interaction dia-

6 International Conference on Deep Excavations (ICDE 2008)


gram. Although structural capacity for compression-dominated failure is not influenced by steel fibre,
which is a characteristic of SFRC based on the foregoing discussions, resistance to bending-
dominated failure increases with both parameters. It is apparent the combination of fct2 and fct3 gives
an enhancement to the structural capacity in the case of low compression.

Figure 7. Influence of fct2 Figure 8. Influence of fct3

Post-cracking flexural strengths do not increase very much with steel fibre dosage. Generally fct2
<0.15f’c and fct3 < fct2. Therefore the increase of structural capacity with the use of steel fibre is lim-
ited, as indicated by Figures 7 and 8. In the case of large bending conventional steel reinforcement is
still required. For SFRC tunnel lining segment, the load cases which will result in high bending espe-
cially in the stages prior to ring closure should be treated with caution, or avoided with appropriate
construction techniques.

5.2 Influence of tensile strain

Steel fibre improves ductility of concrete with limiting tensile strains increased. Effect of tensile strain
parameters, εt1 and εtu, used in the construction of M-N interaction diagram was investigated, and the
results are shown in Figures 9 and 10. Major material parameters adopted in the analyses are as be-
low: f’c = 50 MPa, fct1 = 0.15f’c, fct2 = 0.1f’c and fct3 = 0.8fct2.

Figure 9. Influence of εt1 Figure 10. Influence of εtu

From the two figures it can be seen the tensile strain parameters does not have influence on the inter-
action diagram. Therefore, the increased limiting tensile strains are associated with improved ductility
only, i.e. improved tolerance to deflection or rotation, but does not contribute to the structural capac-
ity.

International Conference on Deep Excavations (ICDE 2008) 7


6 CONCLUSIONS

The structural capacity of tunnel segmental lining was evaluated with the effect of steel fibre taken
into account, based on which comparisons between plain concrete, SFRC, SBRC and SFSBRC were
made and parametric studies performed. As a result, the following findings have been made:

(1) Steel fibre enhances resistance of concrete to bending-dominated failure. For compression-
dominated failure, steel fibre has no effect. Therefore, steel fibre is generally more contributive for
tunnel lining segments in the transportation and installation stages than in the service stage.

(2) Steel fibre can be used together with conventional steel bar to reinforce concrete segment. With
the addition of steel fibre, the capacity of conventional reinforced concrete to resist bending-
dominated failure is further increased, therefore the quantity of steel bar may be reduced.

(3) First crack flexural strength of SFRC does not have influence on structural capacity. The en-
hancement of resistance to bending is due to post-cracking flexural strengths. However, the enhance-
ment is limited because of the relatively low flexural strengths. In the case of large bending conven-
tional steel bar has to be used.

(4) Limiting tensile stains are increased with the addition of steel fibre. This improves the ductility of
concrete but does not contribute to structural capacity.

REFERENCES

German Concrete Association 1992. DBV-Recommendation: design principles of steel fibre reinforced concrete
for tunnelling works.
Kooiman, A.G., Van der Veen, C. & Djorai, M.H. 1998. Steel fibre reinforced concrete (SFRC) tunnel segments
suitable for application in the Second Heinenoord Tunnel, Proceedings of the XIII Congress on challenges
for concrete in the next millennium, Amsterdam, the Netherlands, May 1998: 719-722.
Moyson, D. 1994. Steel fibre reinforced concrete (SFRC) for tunnel linings: A technical approach, Tunnelling
and Ground Conditions, Abdel Salam (ed.) 1994, Rotterdam: Balkema: 673-679.
Sorelli, L. & Toutlemonde, F. 2005. On the design of steel fibre reinforced concrete tunnel lining segments, 11th
International Conference on Fracture, Turlin, Italy, March 20-25, 2005

8 International Conference on Deep Excavations (ICDE 2008)

You might also like